|
|
||||||||
1Baylor College of Medicine, Department of Family and Community Medicine, Texas Medical Center, Houston, Texas 77098; and 2Department of Medicine and Department of Physiology and Biophysics and the Cardiovascular Research Group, The University of Calgary, Calgary, Alberta, Canada T2N 4N1
Submitted 4 February 2004 ; accepted in final form 30 April 2004
| ABSTRACT |
|---|
|
|
|---|
dogs; diastole physiology; systole physiology; pericardium physiology; ventricular function; ventricular pressure; atrial function; atrial pressure
Immediately before atrial contraction, both the atrium and ventricle are relaxed and the pressure gradients and flows across the atrioventricular valves are small. Because the movement of the heart at end diastole is minimal, measurement of pericardial pressure is not affected by the motion of blood or myocardial activation. It has been shown that pericardial pressure is relatively uniform throughout the heart at end diastole (15, 16, 21). Recent investigations in this laboratory have shown that the myocardium plays a relatively minor role in the development of right atrial (RA) and right ventricular (RV) end-diastolic cavitary pressure (PRA and PRV, respectively), with local pericardial pressure being dominant (16, 26). Measurements of early diastolic and pansystolic pericardial pressure have not been reported to date, perhaps due to limitations in transducer technology.
Goto et al. (4) found that pericardial strains over the RV and RA differed significantly during systole and that changes in strain over the RV paralleled changes in RV volume, whereas changes in strain over the RA varied reciprocally with RV volume. These findings suggested that RA and RV pericardial pressure (PpRA and PpRV, respectively) might also be different. The nonlinear stress-strain and creep characteristics of the pericardium (1012, 19) would make estimation of the exact magnitude of these transpericardial stresses from measurements of strain difficult. Accordingly, with recent advances in pericardial pressure measurement technology (7), we undertook to quantify the cycle-specific nonuniformity of pericardial constraint (4).
| METHODS |
|---|
|
|
|---|
After receiving 1020 mg morphine sulfate (im), five mongrel dogs (2025 kg) of both sexes were anesthetized with 12.5 mg/kg thiopental sodium. Anesthesia was maintained with 30 µg·kg1·min1 fentanyl while a 2:1 nitrous oxide-oxygen mixture was delivered by a constant-volume ventilator (model 607, Harvard Apparatus; Millis, MA). All dogs received a tidal volume of 15 ml/kg; a positive end-expiratory pressure of 2 cmH2O was applied. The animals were maintained at 37°C using a circulating-water warming blanket and a constant-temperature bath (model FE2, Haake; Berlin, Germany). The ECG was continuously monitored throughout the experiment.
With the dog in the supine position, a midline sternotomy was performed, and 100200 ml of heparinized Ringer lactate solution were infused to maintain normal aortic pressure. The left lateral surface of the pericardium was opened, and the ventricles were delivered through this incision for purposes of instrumentation. The RV free wall segment length (LRV) and RA appendage diameter (DRA) were measured by sonomicrometry (Triton Technology; San Diego, CA), as previously described (14). Two flat liquid-containing balloon transducers (7) were attached loosely to the epicardium with single stay sutures; one was positioned over the RV free wall, and the other was positioned just cephalad to the RA appendage. The RV balloon (3.0 x 3.0 cm) was fabricated from Silastic sheets (compound AR131, Lot No. 0170, Dow-Corning; Midland, MI) and attached to a 70-cm 8-Fr cardiac catheter (Cordis). The RA balloon (1.2 x 1.4 cm) was connected directly to a pressure transducer (model P23 Db, Statham-Gould; Oxnard, CA) through a 20-cm length of Silastic medical grade tubing (1.0 mm inner diameter, 2.2 mm outer diameter, Dow-Corning).
The heart was repositioned into the pericardium, and the pericardial margins were reapproximated using interrupted 3-0 silk sutures spaced 1 cm apart, with care taken to avoid compromising the pericardial volume (20). RV and RA cavitary pressures were measured using 8-Fr micromanometer-tipped catheters with reference lumens (model PR279, Millar Instruments; Houston, TX) inserted through the internal jugular and femoral veins, respectively. Aortic pressure was measured with an 8-Fr liquid-filled catheter (Cordis; Miami, FL) introduced through the femoral artery. Inferior vena caval and pulmonary arterial pneumatic vascular constrictors (In Vivo Metric Systems; Healdsburg, CA) were also positioned on all dogs.
Calibration curves for both balloons were as described before and after each experiment using a pressurized chamber that loaded the balloons with a Silastic membrane (7) in a way that was similar to that developed by McMahon et al. (17). No differences between pre- and postexperiment calibration curves were found in any experiment. Each balloon had a 3-Fr micromanometer-tipped catheter (model PR249, Millar Instruments) positioned internally to provide a high-fidelity measurement of balloon pressure (7). Intraballoon catheter-tip pressure transducers do not sense the artifacts and oscillations generated from catheter motion and are not affected by the frequency-dependent transmission characteristics of fluid-filled tubes. The dynamic response of this high-fidelity balloon system is linear with a unity-gain frequency response to 200 Hz (7).
Conditioned signals (model VR16, Electronics for Medicine; White Plains, NY) were acquired on a computer (personal computer, Compaq Computer; Houston, TX) using a 12-bit analog-to-digital converter (model 2801, Data Translation; Marlboro, MA). The analog signals were scaled using gain and offset amplifiers and then filtered with a seventh-order Cauer elliptic low-pass active antialiasing filter (model 675, breakpoint = 100 Hz, Frequency Devices; Haverhill, MA) before being sampled at 200 Hz. The digitized data were subsequently analyzed on a computer (Packard Bell; Chatsworth, CA) using analytic software developed in our laboratory (CVSOFT, Odessa Computer Systems; Calgary, Alberta, Canada) and statistical and graphics software (SPSS Incorporated; Chicago, IL) running on a multitasking computer (VAX Computer Digital Equipment, VMS operating system).
After preparatory surgery, all animals were stabilized for 1/2 h before any data were recorded. Mean RA pressure was manipulated from 0 to 25 mmHg by adjusting intravascular volume (Ringer lactate solution was infused or blood removed through a large-bore catheter in an external jugular vein) or by manipulating the inferior vena caval or pulmonary arterial pneumatic constrictors. Hemodynamic and sonomicrometer measurements were obtained continuously for 1 min at each volume-load state (i.e., after
2-mmHg incremental increase of mean RA cavitary pressure); each recording interval began with a 20-s control period. The ventilator was stopped at the end-expiratory position for several cardiac cycles during each recording interval.
Data analysis. Only data collected at end expiration were analyzed. Each micromanometer signal was corrected to equal its corresponding pressure, recorded via the liquid-filled catheter. This correction was performed using software that automatically adjusted the mean value of the pressure recorded by the micromanometer-tipped catheter to equal that recorded by its fluid-filled counterpart. This method was completely automatic and therefore eliminated operator input; nonetheless, all pressure corrections were inspected, although no retrospective adjustments were made.
All data were collected over 60-s intervals. The postexperiment analysis involved the extraction of three consecutive end-expiratory cardiac cycles from the sampled data file (5-ms sampling interval). We studied the diastolic interval from the bottom of the y descent (i.e., the early diastolic minimum in RA pressure) to the beginning of the a wave (pre-a wave) and the systolic interval from the bottom of the x descent (i.e., the early systolic minimum immediately after the c wave) to the peak of the v wave (see Fig. 1).
|
To determine the effect of changing epicardial radial stress on the pressure-dimension properties of the RA and RV, local pericardial pressures were subtracted from intracavitary pressures to produce transmural pressure (TM)-dimension relationships (see Figs. 3 and 4). The PRA-TM loop was calculated by subtracting PpRA from PRA and the PRV-TM loop was calculated by subtracting PpRV from PRV.
|
|
| RESULTS |
|---|
|
|
|---|
|
Figure 3 shows typical RA pressure-dimension loops showing PRA-dimension (top), PpRA-dimension (middle), and PRA-TM-dimension (bottom) relationships (mean PRA equaled 5 mmHg). The PRA-TM-dimension loop is counterclockwise (positive work) for the a loop (atrial contraction) and clockwise for the v loop (ventricular contraction).
Figure 4 shows typical RV cavitary pressure-dimension (top), pericardial pressure-dimension (middle), and transmural pressure-dimension (bottom) relationships with a mean PRA of 10 mmHg. Each plot consists of three consecutive superimposed cardiac cycles all going counterclockwise (i.e., performing positive work).
| DISCUSSION |
|---|
|
|
|---|
In that RA and RV diastolic compliances are similar before atrial contraction, the fundamental "advantage" of atrial systole is that it decreases atrial compliance, thus redistributing blood toward the ventricle, completing ventricular filling (1, 16).
This end-diastolic increase in RV transmural pressure is partially achieved by the reduction in ventricular external constraint (i.e., PpRV) because of the pericardial volume made available by the ejecting atrium (5, 9, 13). If the end-diastolic ventricular constraint was allowed to persist into early diastole, the early filling of the ventricle would not be so rapid, as is the case with pericardial effusions or restrictive cardiac disease (4, 22, 23). This may also be the case in some forms of atrial systolic dysfunction where early diastolic ventricular pericardial constraint is embarrassed by a pathologically distended atrium. It seems interesting that the atrium makes its ejected volume (i.e., pericardial volume) available to the ventricle, yet the converse seems not to be true: in this study, the ejecting ventricle does not seem to increase atrial systolic dimensions.
During ventricular systole, the reservoir function of the atrium permits almost-continuous venous return to the heart (9, 24). The increase in RA volume (1) with a commensurate decrease in RV epicardial constraint (PpRV) permits the atrial storage of blood during systole and facilitates the rapid filling of the RV during early diastole. The atrium is permitted to act as a reservoir during systole, but apparently without impeding early diastolic filling, because PpRV is not increased. Recently, Kilner et al. (9) elegantly described how momentum is conserved during systole as blood accumulates in the atrium. Minimizing ventricular pericardial constraint while maximizing atrial pericardial constraint would seem to maximize the early diastolic atrioventricular valve gradient, as Goto and LeWinter (4) suggested earlier. Without this systolic atrioventricular decoupling, a high PRV at early diastole could embarrass ventricular filling, as seen with pericardial tamponade.
Goto and LeWinter (4) found that atrial pericardial stress-strain areas were almost invariant throughout the cardiac cycle with high PRA. This seems to contrast with our findings showing that the largest fluctuations of pericardial pressure (both during diastole and during systole) occur with maximum volume load. This discrepancy is most probably due to the pericardium being loaded to the noncompliant (high volume) portion of its distinctly nonlinear stress-strain relationship, and therefore large strain fluctuations would not be expected with large stress fluctuations. This suggests that pericardial strain may not be a good indicator of transpericardial radial stress because of the nonlinear stress-strain characteristics of the pericardium (11, 12). Regardless, we find substantial RA systolic cavitary strains with normal mean cavitary pressures (Fig. 3); therefore, pericardial strain is a poor indicator of cavitary volume.
As shown in Fig. 3, the PRA-dimension loops measured in dogs are similar to those found in humans by Nagano et al. (18). In contrast, however, the RA transmural pressure (PRA PpRA)-dimension loop (Fig. 3, bottom) shows the extremely small transmural pressures developed during the v loop. These small transmural pressures during systole and early diastole were consistent throughout all volume states and all animals. (Among animals, the shapes of the a loops were somewhat different, but systolic decoupling was always seen.) These small transmural pressures are notable when one considers the magnitude and range of cavitary pressures and dimension measured during this interval. Nagano et al. (18) and Arakawa et al. (1) measured left atrial compliance using biplane cineangiography to determine volume and the Brockenbrough technique to measure pressure. They found the diastolic left ventricular cavitary compliance to be two to three times larger than the left atrial compliance. We have previously shown that the end-diastolic transmural compliance of the RA in dogs and humans was significantly greater than its RV counterpart (16), which is consistent with the data shown here. To date, this difference in cavitary and transmural pressure-volume relationships has never been examined over the whole cardiac cycle.
Figure 4 demonstrates the importance of accounting for pericardial pressure when assessing the work that the RV does on the blood. Pericardial pressure can be a substantial fraction of RV intracavitary pressure during systole as well as during diastole (6). Furthermore, any attempt to measure RV end-systolic elastance (25) from the intracavitary pressure-dimension loop would be significantly affected by the upward displacement of the loop due to pericardial pressure. Figure 4 also illustrates the value of using high-fidelity pericardial balloons to calculate the transmural pressure of any cardiac cavity in that motion artifacts are absent.
In conclusion, PpRA and PpRV are closely coupled during diastole; however, during RV systole, they become decoupled in that PpRA rises even though PpRV decreases. This regional difference in pericardial constraint during systole becomes more pronounced with increasing PRA.
The end-systolic decrease in ventricular pericardial constraint with concurrent increases in atrial pericardial constraint promotes the early diastolic filling of the ventricle.
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. R. Mitchell, W. A. Whitelaw, R. Sas, E. R. Smith, J. V. Tyberg, and I. Belenkie RV filling modulates LV function by direct ventricular interaction during mechanical ventilation Am J Physiol Heart Circ Physiol, August 1, 2005; 289(2): H549 - H557. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |